Bacterial vaginosis (BV) is the most prevalent cause of vaginal discharge in the U.S. and is associated with serious obstetrical, gynecological, and public health complications including increased risk for acquisition and transmission of STD/HIV. Control of BV has been advocated for decreasing the prevalence of these complications, however, the etiology of BV remains unknown and current treatment regimens are inadequate. Epidemiologic evidence strongly suggests that BV is acquired through sexual activity, yet since the causative agent is not clearly established, investigations into the exact role of sexual activiy in the etiology of BV are hampered. It is clear that Gardnerella vaginalis plays a role in this syndrome since it is present in high concentrations in all women with BV. Newer investigations have highlighted the potential role of uncultivated organisms in BV. There is new data to suggest that BV is a biofilm community and it is possible that some cases of BV are simply due to G vaginalis while others are caused by a more polymicrobic community of organisms. Alternatively, one of the uncultivated organisms could be a primary pathogen. In terms of previous clinical trials examining the benefit of treatment of the male partner, studies conducted in the 1980's either failed to show a benefit of treatment of the male partner in recurrences of BV in the female or showed marginal benefit. However, these studies used short-course therapy (which is only 50% effective or women with BV) and had no measures of compliance or even direct interaction with the male partner. In light of continuing data implicating sexual activity a a risk factor for BV, we propose to conduct a randomized placebo controlled trial of treatment of the male partners of women with recurrent BV to see if treatment of the male decreases recurrence in the female index case. In addition we will collect genital specimens from the couples to look for concordance of organisms especially G vaginalis and archive these samples for future use in determining the prevalence and concordance of uncultivatable organisms associated with BV.